30 Days of Triumph and Turbulence: Exploring a chatbot advertising business idea

Hi, this is Sungho. Over the past month, I've embarked on an intense journey, developing and navigating a promising business idea from inception to conclusion. Join me as I share the highs, lows, and invaluable lessons learned along the way. 1. The Initial Spark The idea seemed simple: leverage the high traffic on GPTs, which are increasingly being used as alternatives to search engines, to generate revenue through text and hyperlink ads. GPTs were rapidly gaining traction, attracting significant user traffic. Where there's traffic, there's potential for advertising revenue. The concept was to embed text and hyperlink ads within GPT interactions, creating a new revenue stream. The primary concern was whether OpenAI would block such initiatives, but it seemed unlikely at the time. 2. Initial Target: Advertisers Identifying the ideal customers was the first step. We targeted media companies, AI SaaS companies, and online course sellers. Through cold emails, we secured initial interest from two advertisers in these sectors. However, challenges quickly emerged when dealing with larger advertisers. For example, one of the largest media companies rejected our proposal due to security policies against AI tools, copyright issues, and uncertainties about OpenAI's advertising policies. It became clear that onboarding large corporations would be difficult due to their strict internal policies. 3. Expanding to GPT Publishers We then turned our focus to collaborating with GPT publishers. Potential partners included some publishers around the top 30 traffic in custom GPTs. However, collecting publisher emails was time-consuming, so we hired freelancers for the data entry. Only some showed genuine interest, confirming a market rate of CPC $0.5+. Many publishers were hesitant due to potential repercussions from OpenAI, and those who were revenue-driven were more open to higher offers from competitors. 4. Pivoting to a Niche Market Realizing the limitations with GPT publishers, we shifted our strategy to character chatbots, which had significant traffic and a stronger desire for monetization. This niche market was easier to attract advertisers in. 5. Engaging Character Chatbot Platforms We reached out to key players in the character chatbot space and held Zoom meetings with character chat platforms, including one of the largest platforms. However, though they liked the concept, they did not firmly commit to using our service, citing concerns about platform compatibility with ads and internal resource allocation. Some showed interest more as an advertiser rather than incorporating ads on their platform. We struggled to find Product-Market Fit. Initially, our landing page was similar to a typical SaaS, but it evolved into a service specializing in advertising products that users of character chatbot platforms would like, such as games, NSFW content, and manga. We contacted advertisers in related industries, including iGaming, VPNs, and manga, as well as marketing agencies. Although we received positive feedback and funding from a few companies, the results did not meet our expectations, and we mutually agreed to terminate the partnerships. 6. Situation Before Halting the Project Before halting the project, we had some traction. Over 700,000 GPT chats were conducted. Our advertisers included casino companies, a Udemy course seller (who used our coupons), and AI SaaS companies. 7. Reasons for Stopping The decision to stop the project was driven by several factors: - Dependency on OpenAI's Policies: The business model heavily relied on OpenAI's policies, creating significant instability. - Lack of User Data: Advertisers lacked sufficient information about GPT users, making it hard to convince them without demographic data. - Reliance on Major Publishers: Securing support from major publishers was challenging. - Traffic Limitations: Only a limited number of character chatbot platforms and GPT publishers had significant traffic. Additionally, due to their competitive landscape, character chatbot platforms were not willing to provide any type of ads that could potentially interrupt user experience. They were still focusing solely on user growth. 8. Reflection Despite the promising start and some initial traction, the project faced insurmountable hurdles due to dependency on a single company's policies and challenges in scaling the business. This experience provided valuable insights into the complexities of integrating ads into emerging digital platforms and highlighted the importance of having control over user data and platform policies. Moving forward, any similar ventures would need to address these fundamental issues to succeed. This journey, while challenging, offered a deep understanding of the digital advertising landscape within the context of AI-driven platforms and the strategic pivots necessary when initial plans encounter roadblocks.